CyclOps Autumn School

What was the Autumn School about?

The CyclOps Autumn School on Knowledge-based Systems explored Knowledge Graphs, Virtual Knowledge Graphs, representation learning, and knowledge-driven AI. The programme combined theoretical foundations with practical demonstrations, covering topics from semantic data access and embeddings to Transformers, LoRA, and ontology-driven information extraction.

– – – –

Content and objective

The programme introduced the foundations of Knowledge Graphs and Virtual Knowledge Graphs, including ontologies and data mappings for heterogeneous data access. It then explored practical applications through demonstrations of Ontopic Suite and openEO + SPARQL, as well as representation learning with Knowledge Graphs. The second day focused on expanding knowledge through latent representations, Transformers and LoRA, LLM-based information extraction, and knowledge-driven data and AI workflows in CyclOps.

– – – –

The three programme areas

• Building Knowledge: From Knowledge Graphs to Virtual Knowledge Graphs – Introduction to Knowledge Graphs, Virtual Knowledge Graphs for data access, ontologies, and data mappings.
• Using Knowledge: From Semantic Access to Representation Learning – Building and consuming Virtual Knowledge Graphs with Ontopic Suite, openEO + SPARQL, and representation learning with Knowledge Graphs.
• Expanding Knowledge: From Latent Representations to Knowledge-Driven AI – Latent spaces, Transformers and LoRA, LLMs in ontology-driven information extraction, and knowledge-driven data and AI workflows in CyclOps.

– – – –

Audience and language

The Autumn School was intended for Master’s students, PhD candidates, researchers, and professionals interested in Knowledge-based Systems, Knowledge Graphs, Virtual Knowledge Graphs, and AI-related topics covered in the programme.

The official language of the Autumn School was English.